How to sum up unique values using case statements having certain conditions - sql

I have a table that may have the same item but with different sizes, I would like to count those items with more than one size (e.g. marvel shirt with S, M sizes will count as "1"), but still be able to count how many S and M. I have 2 outcomes I would like to get. Please see below for more detail.
TABLE B
ITEM_NO ITEM
=========================
3130C MARVEL_SHIRT
1845C SPONGEBOB_BOXERS
A900C CK_COAT
A988C RIDER_JEANS
TABLE C
ITEM_NO SIZE
===============
3130C S
3130C M
1845C M
A900C L
A988C M -
I tried just counting it but it is incorrect since it counts how many distinct sizes are there
select (case substr(item_no, 5, 1)
when 'C' then 'clothes'
when 'T' then 'toys'
else 'misc'
end) inv,
count(item_no) total
,sum (case when C.size = 'S' then 1 else 0 end) AS small
,sum (case when C.size = 'M' then 1 else 0 end) AS med
,sum (case when C.size = 'L' then 1 else 0 end) AS large
,count (distinct C.size) AS multiple_sizes
from B left outer join C on B.item_no = C.item_no
group by substr(item_no, 5, 1);
Actual outcome (incorrect):
INV TOTAL Small Med Large Multiple_Sizes
==========================================================
CLOTHES 4 1 3 1 3
Desired/expected outcome:
INV TOTAL Small Med Large Multiple_Sizes
==========================================================
CLOTHES 4 1 3 1 1
Below is another possible desired outcome given this scenario: What if those with multiple sizes shouldn't be counted separately (i.e. Marvel shirt has multiple sizes, thus it won't count the S or M since it's already counted under Multiple_Sizes)?
INV TOTAL Small Med Large Multiple_Sizes
==========================================================
CLOTHES 4 0 2 1 1

You probably need to group twice (1) by item number (2) by item category:
SELECT SUBSTR(item_no, 5, 1) AS category
, COUNT(*) AS count_products
, SUM(count_small) AS small
, SUM(count_med) AS med
, SUM(count_large) AS large
, SUM(CASE WHEN count_small + count_med + count_large > 1 THEN 1 END) AS has_multiple
FROM (
SELECT b.ITEM_NO
, COUNT(CASE WHEN c.SIZE = 'S' THEN 1 END) AS count_small
, COUNT(CASE WHEN c.SIZE = 'M' THEN 1 END) AS count_med
, COUNT(CASE WHEN c.SIZE = 'L' THEN 1 END) AS count_large
FROM b
LEFT JOIN c ON b.item_no = c.item_no
GROUP BY b.ITEM_NO
) x
GROUP BY SUBSTR(item_no, 5, 1)
| category | count_products | small | med | large | has_multiple |
| C | 4 | 1 | 3 | 1 | 1 |
And the variation:
SELECT SUBSTR(item_no, 5, 1) AS category
, COUNT(*) AS count_products
, SUM(CASE WHEN count_small + count_med + count_large = 1 THEN count_small END) AS small
, SUM(CASE WHEN count_small + count_med + count_large = 1 THEN count_med END) AS med
, SUM(CASE WHEN count_small + count_med + count_large = 1 THEN count_large END) AS large
, SUM(CASE WHEN count_small + count_med + count_large > 1 THEN 1 END) AS has_multiple
FROM (
SELECT b.ITEM_NO
, COUNT(CASE WHEN c.SIZE = 'S' THEN 1 END) AS count_small
, COUNT(CASE WHEN c.SIZE = 'M' THEN 1 END) AS count_med
, COUNT(CASE WHEN c.SIZE = 'L' THEN 1 END) AS count_large
FROM b
LEFT JOIN c ON b.item_no = c.item_no
GROUP BY b.ITEM_NO
) x
GROUP BY SUBSTR(item_no, 5, 1)
| category | count_products | small | med | large | has_multiple |
| C | 4 | 0 | 2 | 1 | 1 |

--creando tabla
create table #temp (itemId int, size nvarchar(1))
--insertando valores
insert into #temp values (1,'S')
insert into #temp values (1,'M')
insert into #temp values (2,'M')
insert into #temp values (3,'L')
insert into #temp values (4,'M')
-- table of Different Item Codes
select
itemId
into #masDeUnItem
from
(select itemId,size from #temp group by itemId,size) t1
group by itemId
having count(1) > 1
-- Variable of Counting different Items
declare #itemsDistintos int
-- Providing Value to Variable
select #itemsDistintos = count(1) from
(
select * from #masDeUnItem
) t1
--Outcome 1
select count(distinct(itemId)) TOTAL
,
sum(case when size = 'S' then 1 else 0 end) SMALL
, sum(case when size = 'M' then 1 else 0 end) MEDIUM
, sum(case when size = 'L' then 1 else 0 end) LARGE
, #itemsDistintos as Multiple_Sizes
from #temp
--Outcome 2
select count(distinct(a.itemId)) TOTAL
,
sum(case when size = 'S' and b.itemId is null then 1 else 0 end) SMALL
, sum(case when size = 'M' and b.itemId is null then 1 else 0 end) MEDIUM
, sum(case when size = 'L' and b.itemId is null then 1 else 0 end) LARGE
, #itemsDistintos as Multiple_Sizes
from #temp a
left join #masDeUnItem b
on a.itemId = b.itemId

Related

Adding a dummy identifier to data that varies by position and value

I am working on a project in SQL Server with diagnosis codes and a patient can have up to 4 codes but not necessarily more than 1 and a patient cannot repeat a code more than once. However, codes can occur in any order. My goal is to be able to count how many times a Diagnosis code appears in total, as well as how often it appears in a set position.
My data currently resembles the following:
PtKey
Order #
Order Date
Diagnosis1
Diagnosis2
Diagnosis3
Diagnosis 4
345
1527
7/12/20
J44.9
R26.2
NULL
NULL
367
1679
7/12/20
R26.2
H27.2
G47.34
NULL
325
1700
7/12/20
G47.34
NULL
NULL
NULL
327
1710
7/12/20
I26.2
J44.9
G47.34
NULL
I would think the best approach would be to create a dummy column here that would match up the diagnosis by position. For example, Diagnosis 1 with A, and Diagnosis 2 with B, etc.
My current plan is to rollup the diagnosis using an unpivot:
UNPIVOT ( Diag for ColumnALL IN (Diagnosis1, Diagnosis2, Diagnosis3, Diagnosis4)) as unpvt
However, this still doesn’t provide a way to count the diagnoses by position on a sales order.
I want it to look like this:
Diagnosis
Total Count
Diag1 Count
Diag2 Count
Diag3 Count
Diag4 Count
J44.9
2
1
1
0
0
R26.2
1
1
0
0
0
H27.2
1
0
1
0
0
I26.2
1
1
0
0
0
G47.34
3
1
0
2
0
You can unpivot using apply and aggregate:
select v.diagnosis, count(*) as cnt,
sum(case when pos = 1 then 1 else 0 end) as pos_1,
sum(case when pos = 2 then 1 else 0 end) as pos_2,
sum(case when pos = 3 then 1 else 0 end) as pos_3,
sum(case when pos = 4 then 1 else 0 end) as pos_4
from data d cross apply
(values (diagnosis1, 1),
(diagnosis2, 2),
(diagnosis3, 3),
(diagnosis4, 4)
) v(diagnosis, pos)
where diagnosis is not null;
Another way is to use UNPIVOT to transform the columns into groupable entities:
SELECT Diagnosis, [Total Count] = COUNT(*),
[Diag1 Count] = SUM(CASE WHEN DiagGroup = N'Diagnosis1' THEN 1 ELSE 0 END),
[Diag2 Count] = SUM(CASE WHEN DiagGroup = N'Diagnosis2' THEN 1 ELSE 0 END),
[Diag3 Count] = SUM(CASE WHEN DiagGroup = N'Diagnosis3' THEN 1 ELSE 0 END),
[Diag4 Count] = SUM(CASE WHEN DiagGroup = N'Diagnosis4' THEN 1 ELSE 0 END)
FROM
(
SELECT * FROM #x UNPIVOT (Diagnosis FOR DiagGroup IN
([Diagnosis1],[Diagnosis2],[Diagnosis3],[Diagnosis4])) up
) AS x GROUP BY Diagnosis;
Example db<>fiddle
You can also manually unpivot via UNION before doing the conditional aggregation:
SELECT Diagnosis, COUNT(*) As Total Count
, SUM(CASE WHEN Position = 1 THEN 1 ELSE 0 END) As [Diag1 Count]
, SUM(CASE WHEN Position = 2 THEN 1 ELSE 0 END) As [Diag2 Count]
, SUM(CASE WHEN Position = 3 THEN 1 ELSE 0 END) As [Diag3 Count]
, SUM(CASE WHEN Position = 4 THEN 1 ELSE 0 END) As [Diag4 Count]
FROM
(
SELECT PtKey, Diagnosis1 As Diagnosis, 1 As Position
FROM [MyTable]
UNION ALL
SELECT PtKey, Diagnosis2 As Diagnosis, 2 As Position
FROM [MyTable]
WHERE Diagnosis2 IS NOT NULL
UNION ALL
SELECT PtKey, Diagnosis3 As Diagnosis, 3 As Position
FROM [MyTable]
WHERE Diagnosis3 IS NOT NULL
UNION ALL
SELECT PtKey, Diagnosis4 As Diagnosis, 4 As Position
FROM [MyTable]
WHERE Diagnosis4 IS NOT NULL
) d
GROUP BY Diagnosis
Borrowing Aaron's fiddle, to avoid needing to rebuild the schema from scratch, and we get this:
https://dbfiddle.uk/?rdbms=sqlserver_2019&fiddle=d1f7f525e175f0f066dd1749c49cc46d

How to Replace NULL Value with 0 (Zero)?

I've just got myself stuck with some SQL query and I'm quite new on this.
I'm using pivot in my query.
This is my SELECT query:
SELECT *
FROM
(SELECT lg.domainNameID AS [Domain ID], COUNT(lg.domainNameID) AS [Fix Count]
FROM tbl_ATT_Request r
INNER JOIN tbl_ATT_Login lg ON lg.workdayID = r.workdayID
WHERE r.requestCategoryID = 1
GROUP BY lg.domainNameID) slct
and this is the output:
Domain | Fix Count
-------+-----------
1 1
2 1
4 2
5 1
And this is my query with PIVOT.
SELECT *
FROM
(SELECT lg.domainNameID AS [Domain ID], COUNT(lg.domainNameID) AS [Fix Count]
FROM tbl_ATT_Request r
INNER JOIN tbl_ATT_Login lg ON lg.workdayID = r.workdayID
WHERE r.requestCategoryID = 1
GROUP BY lg.domainNameID) slct
PIVOT
(SUM(slct.[Fix Count])
FOR slct.[Domain ID] IN ([1],[2],[3],[4],[5])
) AS pvt
This is the output:
1 | 2 | 3 | 4 | 5
1 1 NULL 2 1
Now my problem is how can I replace the NULL values with 0.
Just use conditional aggregation:
SELECT SUM(CASE WHEN Domain_Id = 1 THEN Fix_Count ELSE 0 END) as d_1,
SUM(CASE WHEN Domain_Id = 2 THEN Fix_Count ELSE 0 END) as d_2,
SUM(CASE WHEN Domain_Id = 3 THEN Fix_Count ELSE 0 END) as d_3,
SUM(CASE WHEN Domain_Id = 4 THEN Fix_Count ELSE 0 END) as d_4,
SUM(CASE WHEN Domain_Id = 5 THEN Fix_Count ELSE 0 END) as d_5
FROM (SELECT lg.domainNameID AS Domain_ID, COUNT(*) AS Fix_Count
FROM tbl_ATT_Request r JOIN
tbl_ATT_Login lg
ON lg.workdayID = r.workdayID
WHERE r.requestCategoryID = 1
GROUP BY lg.domainNameID
) d

SQL Select with joins and calculate percentage

I have three tables and I'm trying to make a select statement to give me a result like the one below
Teams:
ID Name
1 A
2 B
3 C
Players:
ID Name TeamID
1 P1 1
2 P2 1
3 P3 2
Goals: (goaltype: H for home, A for away, T for training)
ID PID goaltype
1 1 A
2 1 A
3 1 H
4 2 A
5 2 H
6 3 A
7 3 T
Result will be Like:
Team totalGoals home away trainig percentage[(home/total)*100]
A 5 2 3 0 40%
B 2 0 1 1 0
C 0 0 0 0 0
This is my current query:
select t.name,
count(g.id) as totalGoals,
sum(case when g.GTYPE = 'H' then 1 else 0 end) as home,
sum(case when g.GTYPE = 'A' then 1 else 0 end) as away,
sum(case when g.GTYPE = 'T' then 1 else 0 end) as training,
--(home/totalGoals) as percentage
from teams t
left join players p on p.TeamID = t.id
left join goals g on g.pid = p.id
group by t.name
You can use conditional aggregation to get the results you want:
SELECT t.Name AS Team,
COUNT(g.goaltype) AS totalGoals,
SUM(CASE WHEN g.goaltype = 'H' THEN 1 ELSE 0 END) AS home,
SUM(CASE WHEN g.goaltype = 'A' THEN 1 ELSE 0 END) AS away,
SUM(CASE WHEN g.goaltype = 'T' THEN 1 ELSE 0 END) AS training,
CASE WHEN COUNT(g.goaltype) = 0 THEN 0
ELSE 100.0 * SUM(CASE WHEN g.goaltype = 'H' THEN 1 ELSE 0 END) /
COUNT(g.goaltype)
END AS percentage
FROM Teams t
LEFT JOIN Players p ON p.TeamID = t.ID
LEFT JOIN Goals g ON g.PID = p.ID
GROUP BY t.Name
ORDER BY t.Name
Output:
team totalgoals home away training percentage
A 5 2 3 0 40
B 2 0 1 1 0
C 0 0 0 0 0
Demo on SQLFiddle

Mutually exclusive counts in SQL

The table that I need to query looks like this
ID - Account - Product
1 - 002 - Bike
2 - 003 - Bike
4 - 003 - Motor
5 - 004 - Car
I need to be able to retrieve the number of accounts that purchased the each of the products and combinations of the products, like this
Bike | Car | Motor | Bike&Car | Bike&Motor | Car&Motor | Bike&Car&Motor
Note that an account that purchased a combination of products will be counted as 1.
Please help me in retrieving this data.
You can do this using two levels of aggregation. One method puts the values on different rows:
select has_bike, has_motor, has_car, count(*)
from (select account,
max(case when product = 'bike' then 1 else 0 end) as has_bike,
max(case when product = 'motor' then 1 else 0 end) as has_motor,
max(case when product = 'car' then 1 else 0 end) as has_car
from t
group by account
) t
group by has_bike, has_motor, has_car;
Or in columns:
select sum(has_bike * (1 - has_motor) * (1 - has_car)) as has_only_bike,
sum((1 - has_bike) * has_motor * (1 - has_car)) as has_only_motor,
sum((1 - has_bike) * (1 - has_motor) * has_car) as has_only_car,
. . .
from (select account,
max(case when product = 'bike' then 1 else 0 end) as has_bike,
max(case when product = 'motor' then 1 else 0 end) as has_motor,
max(case when product = 'car' then 1 else 0 end) as has_car
from t
group by account
) t;
If you only have a limited set of Products, then you can use this:
-- Create sample data
CREATE TABLE #tbl(
ID INT,
Account VARCHAR(10),
Product VARCHAR(10)
);
INSERT INTO #tbl VALUES
(1, '002', 'Bike'),
(2, '003', 'Bike'),
(3, '003', 'Motor'),
(4, '004', 'Car');
WITH Cte AS(
SELECT t1.Account, a.Products
FROM #tbl t1
CROSS APPLY (
SELECT STUFF((
SELECT '&' + t2.Product
FROM #tbl t2
WHERE t2.Account = t1.Account
ORDER BY t2.Product
FOR XML PATH(''), type).value('.[1]','nvarchar(max)'),
1, 1, '') AS Products
) a
GROUP BY t1.Account, a.Products
)
SELECT
Bike = SUM(CASE WHEN Products = 'Bike' THEN 1 ELSE 0 END),
Car = SUM(CASE WHEN Products = 'Car' THEN 1 ELSE 0 END),
Motor = SUM(CASE WHEN Products = 'Motor' THEN 1 ELSE 0 END),
[Bike&Car] = SUM(CASE WHEN Products = 'Bike&Car' THEN 1 ELSE 0 END),
[Bike&Motor] = SUM(CASE WHEN Products = 'Bike&Motor' THEN 1 ELSE 0 END),
[Car&Motor] = SUM(CASE WHEN Products = 'Car&Motor' THEN 1 ELSE 0 END),
[Bike&Car&Motor] = SUM(CASE WHEN Products = 'Bike&Car&Motor' THEN 1 ELSE 0 END)
FROM Cte;
DROP TABLE #tbl; -- Remove sample data
The idea is to generate 1 row for each Account, together with a comma-delimited Products. If you execute the query inside the CTE, you will get:
Account Products
---------- ---------------
002 Bike
003 Bike&Motor
004 Car
With that, you can do a conditional aggregation. The above uses a static solution, if you do not know the number of Products, you may need to come up with a dynamic approach.
ONLINE DEMO

Aggregative sum of objects belonging to objects residing inside hierarchy structure

My problem is similar in a way to this one, yet different enough in my understanding.
I have three tables:
Units ([UnitID] int, [UnitParentID] int)
Students ([StudentID] int, [UnitID] int)
Events ([EventID] int, [EventTypeID] int, [StudentID] int)
Students belong to units, units are stacked in a hierarchy (tree form - one parent per child), and each student can have events of different types.
I need to sum up the number of events of each type per user, then aggregate for all users in a unit, then aggregate through hierarchy until I reach the mother of all units.
The result should be something like this:
My tools are SQL Server 2008 and Report Builder 3.
I put up a SQL fiddle with sample data for fun.
Use this query:
;WITH CTE(Id, ParentId, cLevel, Title, ord) AS (
SELECT
u.UnitID, u.UnitParentID, 1,
CAST('Unit ' + CAST(ROW_NUMBER() OVER (ORDER BY u.UnitID) AS varchar(3)) AS varchar(max)),
CAST(RIGHT('000' + CAST(ROW_NUMBER() OVER (ORDER BY u.UnitID) AS varchar(3)), 3) AS varchar(max))
FROM
dbo.Units u
WHERE
u.UnitParentID IS NULL
UNION ALL
SELECT
u.UnitID, u.UnitParentID, c.cLevel + 1,
c.Title + '.' + CAST(ROW_NUMBER() OVER (PARTITION BY c.cLevel ORDER BY c.Id) AS varchar(3)),
c.ord + RIGHT('000' + CAST(ROW_NUMBER() OVER (ORDER BY u.UnitID) AS varchar(3)), 3)
FROM
dbo.Units u
JOIN
CTE c ON c.Id = u.UnitParentID
WHERE
u.UnitParentID IS NOT NULL
), Units AS (
SELECT
u.Id, u.ParentId, u.cLevel, u.Title, u.ord,
SUM(CASE WHEN e.EventTypeId = 1 THEN 1 ELSE 0 END) AS EventA,
SUM(CASE WHEN e.EventTypeId = 2 THEN 1 ELSE 0 END) AS EventB,
SUM(CASE WHEN e.EventTypeId = 3 THEN 1 ELSE 0 END) AS EventC,
SUM(CASE WHEN e.EventTypeId = 4 THEN 1 ELSE 0 END) AS EventD
FROM
CTE u
LEFT JOIN
dbo.Students s ON u.Id = s.UnitId
LEFT JOIN
dbo.[Events] e ON s.StudentId = e.StudentId
GROUP BY
u.Id, u.ParentId, u.cLevel, u.Title, u.ord
), addStudents AS (
SELECT *
FROM Units
UNION ALL
SELECT
s.StudentId, u.Id, u.cLevel + 1,
'Student ' + CAST(s.StudentId AS varchar(3)),
u.ord + RIGHT('000' + CAST(s.StudentId AS varchar(3)), 0),
SUM(CASE WHEN e.EventTypeId = 1 THEN 1 ELSE 0 END),
SUM(CASE WHEN e.EventTypeId = 2 THEN 1 ELSE 0 END),
SUM(CASE WHEN e.EventTypeId = 3 THEN 1 ELSE 0 END),
SUM(CASE WHEN e.EventTypeId = 4 THEN 1 ELSE 0 END)
FROM Units u
JOIN
dbo.Students s ON u.Id = s.UnitId
LEFT JOIN
dbo.[Events] e ON s.StudentId = e.StudentId
GROUP BY
s.StudentID, u.ID, u.cLevel, u.ord
)
SELECT --TOP(10)
REPLICATE(' ', cLevel) + Title As Title,
EventA, EventB, EventC, EventD
FROM
addStudents
ORDER BY
ord
For this:
Title | EventA | EventB | EventC | EventD
-----------------+--------+---------+--------+--------
Unit 1 | 0 | 1 | 0 | 0
Student 6 | 0 | 1 | 0 | 0
Unit 1.1 | 0 | 0 | 0 | 1
Student 21 | 0 | 0 | 0 | 1
Student 33 | 0 | 0 | 0 | 0
Unit 1.1.1 | 0 | 0 | 0 | 0
Student 23 | 0 | 0 | 0 | 0
Unit 1.1.1.1 | 3 | 2 | 3 | 0
Student 10 | 0 | 0 | 0 | 0
Student 17 | 1 | 0 | 0 | 0
...
SQL Fiddle Demo
Do you need also the hierarchy to be sorted / visualized? At least this will calculate the sums, but the order of the data is pretty random :)
;with CTE as (
select S.StudentId as UnitID, S.UnitId as UnitParentID,
S.StudentID, 'Student' as Type
from Students S
union all
select U.UnitId, U.UnitParentId,
CTE.StudentId as StudentID, 'Unit ' as Type
from
Units U
join CTE
on U.UnitId = CTE.UnitParentId
)
select C.Type + ' ' + convert(varchar, C.UnitId),
sum(case when EventTypeId = 1 then 1 else 0 end) as E1,
sum(case when EventTypeId = 2 then 1 else 0 end) as E2,
sum(case when EventTypeId = 3 then 1 else 0 end) as E3,
sum(case when EventTypeId = 4 then 1 else 0 end) as E4
from
CTE C
left outer join events E on C.StudentId = E.StudentId
group by
C.Type, C.UnitId
SQL Fiddle
If you need also the hierarchy to be in order, you'll probably have add few extra CTEs to get the numbering from top down with something like #shA.t did. This gathers the hierarchy separately for each student, so it's not really possible to add the level numbers in a simple way.